JP2019203879A - Multi-spectral gas quantification and differentiation method for optical gas imaging camera - Google Patents

Abstract

To provide an infrared image acquisition method with excellent visualization of gas leaks, short inspection time and high safety.SOLUTION: The method includes: acquiring a multi-spectral image of a detected radiance including a plurality of pixels; estimating a background radiance for the pixels; calculating a gas concentration-length for the pixels on the basis of the detected radiance and the estimated background radiance; and triggering an alert if it exceeds a gas threshold level. A multi-spectral configuration of the multi-spectral optical gas imaging camera includes a reference band that is outside an absorption wavelength range of a target gas, and an active band that includes at least a portion of the absorption wavelength range of the target gas. Estimating the background radiance includes determining a model relating the detected radiance of the active band to the detected radiance of the reference band, and using the model as a calibration model to estimate the background radiance for the active band with or without the presence of a target gas.SELECTED DRAWING: Figure 3

Description

  In recent years, infrared (IR) optical gas imaging (OGI) cameras are superior to conventional gas detection methods (such as detectors using a catalyst) in visualizing gas leakage, and have a shorter inspection time and safety. Because of its high performance, it is widely used for detection and monitoring of gas leaks.

  In addition to practical qualitative applications, quantitative applications have been proposed for IR OGI cameras. For example, by appropriate calibration, the gas concentration and length of the imaged plume (smoke flow) (concentration integrated over the path length, in units of parts per million meters, i.e. ppm m) Can be quantified. The gas concentration / length is also called gas concentration path length (CPL).

  IR OGI is one of the two types of gas leak detection and repair (LDAR) research methods in the US oil and gas industry recognized by the US Environmental Protection Agency (EPA). This is a method according to EPA21.

  This summary is provided to introduce a selection of concepts that are further described below in the detailed description. This summary is not intended to identify key or essential features of the claimed subject matter, nor is it intended to be used as an aid in limiting the scope of the claimed subject matter.

  In one or more aspects of the present invention, a method for quantifying gas concentration / length includes: obtaining a multispectral image of detected radiance including a plurality of pixels using a multispectral optical gas imaging camera; Estimating at least one of the background radiances, calculating at least one gas concentration / length of the plurality of pixels based on the detected radiance and the estimated background radiance, and a plurality of alarm conditions A step of issuing an alarm when each alarm condition in the list is satisfied may be included. The list of alarm conditions may include a gas concentration / length of at least one pixel that exceeds a gas threshold level. The multispectral configuration of the multispectral optical gas imaging camera may include a reference band that is outside the absorption wavelength range of the target gas and an active band that includes at least a portion of the absorption wavelength range of the target gas. Estimating the background radiance includes determining a model that relates the detected radiance of the active band to the detected radiance of the reference band, and using the model as a calibration model in the presence or absence of the target gas. Estimating the background radiance of the active band may be included.

  In one or more aspects of the invention, the gas concentration and length determination system may include a multispectral optical gas imaging camera and a processor connected to the multispectral optical gas imaging camera. The processor reads an image including a plurality of pixels, estimates a background radiance of at least one of the plurality of pixels, a gas concentration of at least one of the plurality of pixels based on the image and the background radiance, An alarm may be issued if the length is calculated and if the gas concentration / length of at least one pixel exceeds the gas threshold level. The multispectral configuration of the multispectral optical gas imaging camera may include a reference band that is outside the absorption wavelength range of the target gas and an active band that includes at least a portion of the absorption wavelength range of the target gas. Estimating the background radiance is to determine a model that relates the detected radiance of the active band to the detected radiance of the reference band, and using the model as a calibration model, in the presence or absence of the target gas. Estimating the background radiance of the active band may be included.

  In one or more aspects of the invention, a non-transitory computer readable medium stores computer readable program code embodied therein, the computer readable program code reading an image including a plurality of pixels, And calculating a gas concentration / length of at least one of the plurality of pixels based on the image and the background radiance, and the gas concentration / length of the at least one pixel is a gas threshold level. If it exceeds, an alarm may be issued. The multispectral configuration of the multispectral optical gas imaging camera may include a reference band that is outside the absorption wavelength range of the target gas and an active band that can include at least a portion of the absorption wavelength range of the target gas. Estimating the background radiance is to determine a model that relates the detected radiance of the active band to the detected radiance of the reference band, and using the model as a calibration model, in the presence or absence of the target gas. Estimating the background radiance of the active band may be included.

  Other aspects and advantages will be apparent from the following description and the appended claims.

  Aspects of the present invention will be described with reference to the accompanying drawings. The accompanying drawings, however, illustrate only certain aspects or implementations of one or more embodiments of the invention by way of example and do not limit the scope of the claims.

1 illustrates a gas concentration and length determination system according to one or more embodiments of the present invention. 1 illustrates a gas concentration and length determination system according to one or more embodiments of the present invention. 1 illustrates a gas concentration and length determination system according to one or more embodiments of the present invention. FIG. 2 is a schematic diagram illustrating a radiative transfer model mechanism for gas detection and quantification according to one or more embodiments of the present invention. 4 is a flowchart of a gas concentration / length determination method according to one or more embodiments of the present invention. ₂ illustrates an exemplary three-band configuration for simulation of methane (CH ₄ ) and water vapor (H ₂ O) transmission spectra and a mid-infrared (MWIR) multi-spectral OGI system according to one or more embodiments of the present invention. FIG. 6 illustrates a simulation of a radiance ratio between an active band and a reference band according to one or more embodiments of the present invention. FIG. 6 illustrates a simulation of a radiance ratio between an active band and a reference band according to one or more embodiments of the present invention. The radiance estimated with the analytical model about two active bands based on the radiance of a reference band is shown. In one embodiment of the present invention, the radiance estimate is compared to a numerical result based on Planck's law. FIG. 6 illustrates a simulation of methane and water vapor discrimination feature signals according to one or more embodiments of the present invention. FIG.

  Hereinafter, specific embodiments of the present invention will be described in detail with reference to the accompanying drawings. Similar components in the Figures are denoted by similar reference numerals for consistency.

  In the following detailed description of embodiments according to the present invention, numerous specific details are set forth in order to provide a thorough understanding of the present invention. However, it will be apparent to one skilled in the art that the present invention may be practiced without these details. In other instances, well-known features are not described in order not to unnecessarily complicate the description.

  Throughout this application, ordinal numbers (eg, first, second, third, etc.) may be used as adjectives for certain components (ie, any nouns in this application). The use of ordinal numbers does not imply or imply a particular order of components, but the use of “before”, “after”, “single”, and other such terms. Unless explicitly disclosed by the above, any component is not limited to only one component. Rather, the use of ordinal numbers distinguishes components. As an example, the first component is different from the second component, and the first component may include two or more components, and is later in the order of the components than the second element ( Alternatively, it may be the first).

Overall, embodiments of the present invention use an image taken with a camera to subtract the background radiance estimated using at least one active band and one reference band of the spectrum of the image from the measured radiance. , Methods, systems, and non-transitory computer-readable media for quantifying gas concentration and length are provided. The spectral range may include any suitable range of wavelengths. In one or more embodiments may be used an infrared (MWIR) within the range of ^{3 × 10 -6 m~5 × 10 -6} m. In one or more embodiments may be used far infrared (LWIR) in the range of ^{7 × 10 -6 m~14 × 10 -6} m. Mid-infrared and far-infrared are useful for hydrocarbon gas detection applications. In one or more embodiments, the camera is an optical gas imaging camera that can sense infrared radiation.

  One or more embodiments of the present invention provide a method of reducing false alarms by distinguishing between a target gas and a contaminated gas.

  FIG. 1A illustrates a gas concentration and length determination system according to one or more embodiments of the present invention. Optical gas imaging (OGI) camera 30 may be connected to one or more computer processors 102. In one or more embodiments, the OGI camera 30 is a multispectral OGI camera. The OGI camera 30 and the computer processor 102 may be connected by wire or may be connected wirelessly. One or more processors may be included in computer system 100.

  Embodiments of the present invention may be implemented using virtually any type of computer system 100, regardless of the platform used. For example, a user computing device can have one or more portable devices (eg, laptop computers, smartphones, personal digital assistants, tablet computers, or other portable devices) for performing one or more embodiments of the present invention. ), A desktop computer, a server, a blade in a server chassis, or any other type of computing device including at least minimal processing power, memory and input / output devices. For example, as shown in FIG. 1A, the computer system 100 includes one or more processors 102, associated memory 104 (eg, random access memory (RAM), cache memory, flash memory, etc.) and one or more storages. Device 106 (eg, hard disk, optical drive (such as a compact disk (CD) drive or digital versatile disk (DVD) drive), flash memory stick, etc.), and many other components and functions may be included. The processor 102 may be an integrated circuit for processing instructions. For example, the processor 102 may be one or more cores or microcores of the processor.

  The computer system 100 may also include one or more input devices 110, such as a touch screen, keyboard, mouse, microphone, touch pad, electronic pen, or any other type of input device. The computer system 100 may also include a screen (eg, a liquid crystal display (LCD), a plasma display, a touch screen, a cathode ray (CRT) monitor, a projector, or other display device), a printer, an external storage device, or any other output. One or more output devices 108, such as devices, may be included. One or more of the output devices 108 may be the same as or different from the input device. The output device 108 may include a light (for example, flashing red light) that can be used as an alarm when the gas concentration / length exceeds the gas threshold level, a warning sound, a buzzer, and the like. The output device may create and / or receive facsimile (fax), e-mail, short message service (SMS) text, and the like. In one or more embodiments, it may indicate a contaminated gas at a level that exceeds a preset species identification threshold and distorts the gas concentration / length measurements, and may refrain from alarming if a false alarm is issued.

  FIG. 1B illustrates that in accordance with one or more embodiments, the computer system 100 can connect to a network 112 (eg, a wide area network (such as a local area network (LAN), the Internet, etc.) via a network interface connection (not shown) WAN), mobile network, or any other type of network). The input device and output device may be locally connected or remotely connected (eg, via a network 112 connected to the processor 102, memory 104, and storage device 106). The OGI camera 30 may be connected to one or more computer processors 102 by a network 112. There are many different types of computer systems, and the above input / output devices may take other forms.

  All or part of the software instructions in the form of computer readable program code for executing the embodiments of the present invention may be temporarily or permanently stored in a CD, DVD, storage device, floppy disk, tape, flash memory, physical memory. Or a non-transitory computer readable medium, such as any other computer readable storage medium. In particular, software instructions may correspond to computer readable program code that, when executed by processor 102, performs one or more embodiments of the invention. Also, all described steps performed by a processor executing software instructions may be in the form of hardware, such as a circuit, in one or more embodiments. Those skilled in the art will appreciate that the hardware may comprise an application specific integrated circuit or other suitable circuitry.

  One or more components of the computer system 100 may be remotely located and connected to other components via the network 112. Furthermore, one or more embodiments of the present invention may be implemented in a distributed system having multiple nodes, and portions of the present invention may be located at different nodes within the distributed system. In one or more embodiments of the invention, a node corresponds to an individual computing device. Alternatively, the node may correspond to a processor having an associated physical memory. Alternatively, a node may correspond to a processor or processor microcore that shares memory and / or resources.

  FIG. 1C illustrates an OGI camera 30 according to one or more embodiments. The OGI camera 30 is connected to the detector 32. This detector may be of the type normally used in residential or industrial environments, for example a thermometer or a hygrometer. The connection may be wired or wireless and may include a connection via a network. In one or more embodiments, the detector 32 may be integral with the OGI camera. In one or more embodiments, the OGI camera 30 may be integrated with an integrated circuit.

  In one or more embodiments, the OGI camera 30 may be used to capture images for quantifying gas concentration and length, particularly from commercial, industrial, marine, residential, or rural environments.

  FIG. 2 shows the imaging of gas leakage by OGI according to one or more embodiments and the gas concentration / length (concentration integrated by thickness or length, for example, parts per million / meter, ie, ppm · 1 shows a schematic diagram of the physical basis of m) quantification (known as a radiative transfer model). The OGI camera 30 may include at least one long wavelength transmission filter that passes at least a part of the absorption band of the target gas. The OGI camera 30 may also be equipped with a reference band long wavelength transmission filter that does not pass through the absorption band of the target gas. The long wavelength transmission filter transmits light exceeding a specific wavelength. In one or more embodiments, the OGI camera 30 may include different types of filters, such as short wavelength transmission filters or band transmission filters. The target gas may be, for example, methane, sulfur hexafluoride, carbon monoxide, and carbon dioxide. One or more of these target gases can be detected, for example, in the environment of the oil and gas industry. However, one or more embodiments of the present invention may be applied to other environments including, for example, residential environments.

As shown in FIG. 2, the radiance 40, L _b (λ) (λ indicates the radiation wavelength) from the background 10 passes through the gas plume 20. The intensity of the background radiance 40 may be weakened by gas absorption. Absorption by the gas plume 20 may be characterized by a transmittance coefficient τ _p (λ), so the transmitted radiance 50 reaching the OGI camera 30 will be L _b (λ) τ _p (λ). The transmittance coefficient τ _p (λ) is related to the gas concentration / length _γ by Lambert-Beer's law: τ _p (λ) = exp ( _−γ α (λ)). Here, the absorbance spectrum α (λ) is unique to each gas species and is known.

On the other hand, in one or more embodiments of the invention, the gas plume 20 that absorbs infrared radiation also radiates thermal energy 60. This is characterized by the product of the blackbody radiation (Planck function) B (T _p , λ) at the gas temperature (T _p ) and the emissivity of the gas (1-τ _p ). Thus, in one or more embodiments, the OGI camera 30 records the background radiance L _b (λ) adjusted by absorption by the gas plume 20 given by equation (1).

  Depending on the relative temperature difference between the gas plume 20 and the background 40, the gas plume 20 may be visualized as dark contrast or light contrast with respect to the background 40 in the image.

Assuming that L _b (λ) and B (T _p , λ) are known in equation (1), the simplified radiative transfer equation (1) determines the permeability coefficient τ _p (λ) and thus gas The physical principle of the concentration / length determination of the plume 20 can also be obtained. B (T _p , λ) is easily obtained by measuring the temperature of the gas plume 20 and calculating it using a Planck function. Therefore, obtaining the background radiance L _b (λ) is an important issue for gas concentration / length determination in optical gas imaging (OGI).

  In one or more embodiments of the present invention, a method for estimating background radiance with optical gas images comprises a multispectral technique using multiple filters and machine learning techniques, and gas concentration / length quantification processing based thereon. It may be adopted. In this method, the background image may be generated analytically.

  Thus, one or more embodiments of the present invention can have the advantage that the background radiance can be estimated in the presence of a gas plume and the gas concentration and length can be quantified. Advantageously, one or more embodiments of the present invention can provide reliable background radiance estimation under a wide range of environmental conditions.

  FIG. 3 is a flowchart describing a method for quantifying gas concentration and length based on background radiance estimation using a multispectral method according to one or more embodiments. In one or more embodiments of the present invention, one or more of the steps shown in FIG. 3 may be omitted or repeated, or additional steps may be performed. Therefore, the scope of the present invention should not be considered limited to the specific sequence of steps shown in FIG.

  In step 310, an image is acquired. This image may or may not contain a plume. In one or more embodiments, this image may be an infrared image acquired by an OGI camera. In order to clarify the explanation and distinguish this image from other images, this image will be referred to as an infrared (IR) image hereinafter. However, this term is not intended to limit the image to infrared. It will be readily appreciated that images of other emission wavelengths may be included. One IR image may be acquired, a series of IR images may be acquired, and a series of video image frames is the same. The brightness of the pixels of the IR image may be calibrated using known calibration data of the OGI camera to indicate either the temperature or radiance value of the scene.

  In step 320, the method estimates the background radiance using an analytical model derived from simulation data and / or experimental measurements. Background radiance estimation may be based on inputs of known atmospheric temperature, background temperature, background reflectivity, and similar environmental parameters. Analytical models for background radiance estimation were derived using machine learning on individual band-integrated radiance simulation data or experimental measurement data from background scenes for various background material thermal properties and a wide range of environmental conditions. May be. The background scene herein refers to a background image that is free of target gas or polluted gas. Thermal properties may include reflectivity, emissivity, and the like. Environmental conditions may include background material temperature, atmospheric temperature, temperature difference between the background material and the atmosphere, sunlight conditions, and the like.

  In step 330, the method calculates the gas concentration / length of each pixel using the infrared image data and the estimated background radiance. In one or more embodiments, the gas temperature may be measured with a thermometer or assumed to be the same as the outside air temperature.

  The calculation of the gas concentration / length may be based on the above radiative transfer model, or may use the procedure disclosed in International Patent Publication No. WO2017104607A1.

  In step 340, the method identifies a target gas and at least one contaminated gas.

  In one or more embodiments, the method described with reference to the flowchart of FIG. 3 repeats steps 310-350 multiple times to process multiple images, such as those that can be generated in a video stream. May be.

  In step 350, the method issues an alarm if the gas concentration / length of at least one pixel exceeds a gas threshold level. The alarm may take many forms and may include multiple forms. One form of alarm may be a colored light (eg, red light) on the surface of the user control board. This light may be lit and / or flashing. The alarm may be a sound, for example, a bell, a siren, a horn or the like. The alert may be an email, fax, short message system (SMS) text message, telephone communication, and the like. The alarm may be displayed on a computer screen, a mobile device, a mobile phone, or the like. An alert may be provided to one or more locations and / or one or more users. The alarm may be transmitted via a wired and / or wireless system or network. The alert destination may be another device that can automatically perform one or more functions in response to a user and / or alert.

4 to 6 show simulation examples of background radiance estimation processing using an analysis model as in step 320 of FIG. FIG. 4 represents a simulation of the transmittance spectrum of methane (CH ₄ ) 440 and water vapor (H ₂ O) 450 in one or more embodiments. Also shown are three long wavelength transmission filters 410, 420, 430 that can be used in a mid-infrared (MWIR) multispectral OGI three-band configuration. The three-band configuration in this case is an example of a multispectral configuration. Those skilled in the art will readily appreciate that a multispectral configuration may include more than one band. Three bands can be constructed using three long wavelength transmission filters. The first active band can be expressed as a band obtained by subtracting the band transmitted through the filter 430 from the band transmitted through the filter 410. Similarly, the second active band can be expressed as a band obtained by subtracting the band transmitted through the filter 430 from the band transmitted through the filter 420. The reference band is a band that passes through the filter 430. The reference band can be represented as being outside the absorption wavelength range of the target gas (methane 440 in this example). In the present disclosure, the term “outside the absorption wavelength range” does not refer to a region where there is no absorption at all, and only a relatively small amount of absorption may occur in that region. Those skilled in the art can easily understand such views.
In one or more embodiments, a background radiance similar to that of FIG. 2 can be simulated as given by equation (2).
L _b (λ) = ρ _b (λ) B (λ, T _a ) + (1−ρ _b (λ)) B (λ, T _b ) (2)
Where ρ _b , T _a , and T _b are the background material reflectivity, atmospheric temperature, and background material temperature, respectively. Thus, background radiance may include contributions of atmospheric radiance reflected from the background and radiance emitted directly from the background. In some examples, the background may be a wall or sky, or any other object.

In single-band or multi-band OGI systems, the radiance detected in the individual bands can be expressed as given by equation (3).
_Where τ _filter is the transmittance spectrum of the filter, s _sensor is the spectral sensitivity of the sensor, and integration is performed over the entire wavelength range of the band. When simulating the detected radiance, the spectral sensitivity of the sensor may be set to 1 for convenience.

5A-5B illustrate exemplary simulations of the radiance ratio L _{off_active} / L _{off_reference} between the active band and the reference band for a range of background temperature T _b and various temperature differences ΔT _ab between ambient air and background material. Provided for two values of background reflectance ρ _b = 0.1 (FIG. 5A) and ρ _b = 0.9 (FIG. 5B). Here, the subscript “off” means that no gas is present. Two active bands, band 1 (510, 511) and band 2 (520, 521) are shown for a range of temperature differences ΔT _ab , with larger temperature differences being shown for each curve 510, 511, 520. , And at the top of 521.

In one or more embodiments, machine learning may be performed on the simulation data as in FIGS. 5A and 5B to derive an analytical model for the radiance of the nth active band and the radiance of the reference band. . Examples of machine learning include regression and neural networks including linear regression. In one or more embodiments, the analytical model may be given by equation (4).
L _{off_active (n)} = [C ₀ + C ₁ · T _b + C ₂ · (ρ _b ΔT _ab )] · L _{off_ref} (4)
In the equation, C0, C1, and C2 are model parameters obtained by machine learning of simulation data.

  FIG. 6 shows two active bands (band 1 (610)) based on the radiance of the reference band compared to numerical results based on Planck's law under different material and environmental conditions used in the learning data. And band 2 (620)) represents the background radiance estimated with the analytical model according to one or more embodiments.

In one or more embodiments, the gas signal parameter ΔL is defined as given in equation (5) for each active band.
ΔL = L _off − L _on (5)
_Where L _off is the background radiance estimated in each active band and L _on is the radiance detected in the active band in the presence of gas. Using the relative ratio ΔL of the two active bands (ie, ΔL _band1 / ΔL _band2 ) as a feature parameter, and comparing this feature parameter with a preset species identification threshold, the target gas and the falsely reported species (ie, contaminating species) May be identified.

  It is necessary to preset the species identification threshold of the identification feature parameter in the OGI system for each related misinformation type. These species discrimination thresholds may be determined by theoretical simulation or experimental calibration.

FIG. 7 shows simulation values ΔL _band1 / ΔL _band2 of methane 740 and water vapor 720 over a wide range of gas concentrations and environmental conditions. In this example, the background temperature is T _b , the temperature of the gas species based on the three-band configuration selected in FIG. 4 is T, and an exemplary species identification threshold that identifies the corresponding methane 740 and water vapor 720. Is 2.4 (760).

  In one or more embodiments of the invention, the gas concentration and length determination system may include an OGI camera 30 connected to the processor 102. The processor 102 performs one or more of a plurality of variations of the gas concentration / length quantification method described above to provide the gas concentration / length, estimate background radiance, and / or reduce false alarms. May be.

  In one or more embodiments, the processor may be incorporated into a computing device. The computing device may be a portable computer device (for example, a smartphone, a tablet computer, a laptop computer, an electronic book terminal, etc.), a desktop personal computer (PC), a kiosk terminal, a server, a main frame, a set top box, or the like. Each computing device may be operated by a user and uses one or more graphical user interfaces (GUIs) to calculate gas concentration / length, background radiance estimate, and / or background temperature estimate. And / or a request from the user may be created to display information to the user. The user request may specify the output location (eg, display device, storage location, printer, etc.) of the calculated data. In one or more embodiments, various elements of the computing device may be combined to create a single element. Similarly, a function performed by one element may be performed by two or more elements.

  In one or more embodiments, the computing device may include multiple computing devices connected to each other.

  In one or more embodiments, the OGI camera may be integrated with the computing device.

  One or more images acquired with an infrared (IR) optical gas imaging (OGI) camera may be transferred to at least one of a plurality of user computing devices for processing and storage.

  In one or more embodiments, the gas concentration and length determination system may include a predetermined background radiance calculation model. This model may be stored in a camera or computer system, accessed through a network or input device, or made available to the system by any other means.

  In one or more embodiments, the gas concentration and length quantification system uses a predetermined background radiance calculation model, and based on environmental and / or background material parameters and radiance measured within a reference band. The background radiance of the active band may be calculated in the presence or absence of gas. The user may enter parameters.

  All or part of the software instructions in the form of computer readable program code for executing the embodiments of the present invention may be temporarily or permanently stored in a CD, DVD, storage device, floppy disk, tape, flash memory, physical memory. Or a non-transitory computer readable medium, such as any other computer readable storage medium. Specifically, the software instructions may correspond to computer readable program code that, when executed by a processor, performs one or more embodiments of the invention. Also, all contemplated processes performed by a processor executing software instructions may be in the form of hardware, such as a circuit, in one or more embodiments. Those skilled in the art will appreciate that the hardware may be comprised of an application specific integrated circuit or other suitable circuitry.

  Furthermore, one or more components of the arithmetic device may be arranged in a remote place and connected to other components via a network. Further, one or more embodiments of the present invention may be implemented in a distributed system having multiple nodes, and portions of the present invention may be located at different nodes within the distributed system. . In one or more embodiments of the invention, a node corresponds to an individual computing device. Alternatively, the node may correspond to a processor having an associated physical memory. Alternatively, a node may correspond to a processor or processor microcore that shares memory and / or resources.

  In one or more embodiments, a non-transitory computer readable medium (CRM) may store computer readable program code embodied therein. The computer readable program code reads an image containing a plurality of pixels, estimates background radiance, calculates gas concentration / length, and issues an alarm if the gas concentration / length exceeds a gas threshold level. Estimating the background radiance includes receiving environmental parameters and background material parameter inputs, and using a background radiance calculation model stored in a non-transitory computer readable medium, And calculating the background radiance of the active band in the presence or absence of the gas of interest based on the radiance measured in the reference band.

  Although one or more embodiments of the present invention have been described with respect to the infrared portion of the spectrum, one of ordinary skill in the art will readily appreciate that the methods disclosed in the present invention can be applied to radiation of other wavelengths.

  Although the present invention has been described with respect to a limited number of embodiments, those skilled in the art having the benefit of this disclosure will appreciate that other embodiments can be devised without departing from the scope of the invention disclosed herein. Accordingly, the scope of the invention should be limited only by the attached claims.

In one or more aspects of the present invention, the processing in the program for executing a predetermined processing on the computer, captured by multispectral optical gas imaging camera, reading an image including a plurality of pixels Rukoto, a plurality of pixels At least one of estimating the background radiance, calculating at least one gas the concentration-length of the plurality of pixels based on the image and background radiance, and gas concentration, length of at least one pixel of the exceed the gas threshold level may include that originating the alarm. The multispectral configuration of the multispectral optical gas imaging camera may include a reference band that is outside the absorption wavelength range of the target gas and an active band that can include at least a portion of the absorption wavelength range of the target gas. Estimating the background radiance is to determine a model that relates the detected radiance of the active band to the detected radiance of the reference band, and using the model as a calibration model, in the presence or absence of the target gas. Estimating the background radiance of the active band may be included.

In one or more embodiments, the processor may be incorporated into a computing device. Arithmetic devices are portable computer devices (for example, smart phones, tablet computers, laptop computers, electronic book terminals, etc.), desktop personal computers (PCs), kiosks (registered trademark) terminals, servers, mainframes, set top boxes, etc. Also good. Each computing device may be operated by a user and uses one or more graphical user interfaces (GUIs) to calculate gas concentration / length, background radiance estimate, and / or background temperature estimate. And / or a request from the user may be created to display information to the user. The user request may specify the output location (eg, display device, storage location, printer, etc.) of the calculated data. In one or more embodiments, various elements of the computing device may be combined to create a single element. Similarly, a function performed by one element may be performed by two or more elements.

Claims (23)

A method for quantifying gas concentration and length,
Obtaining a multispectral image of detected radiance comprising a plurality of pixels using a multispectral optical gas imaging camera;
Estimating a background radiance of at least one of the plurality of pixels;
Calculating the gas concentration / length of the at least one of the plurality of pixels based on the detected radiance and the estimated background radiance; and the gas concentration / length of at least one pixel exceeding a gas threshold level Including a step of issuing an alarm when each alarm condition in a list of a plurality of alarm conditions is satisfied,
The multispectral configuration of the multispectral optical gas imaging camera includes a reference band that is outside an absorption wavelength range of a target gas and an active band that includes at least a portion of the absorption wavelength range of the target gas;
The step of estimating the background radiance includes
Determining a model relating the detected radiance of the active band to the detected radiance of the reference band; and using the model as a calibration model, the background radiance of the active band in the presence or absence of the target gas Estimating the method.   The method of claim 1, wherein the image is an infrared image and is acquired by the multispectral optical gas imaging camera capable of sensing infrared radiation.   The method of claim 1, wherein the model is an analytical model determined using machine learning including regression and neural networks. The multispectral configuration includes a plurality of active bands;
The plurality of active bands and the reference band are determined using a plurality of long wavelength transmission filters, one more than the number of active bands,
Each of the plurality of active bands includes the reference band;
The plurality of active bands include a first active band, and the first active band includes a first long wavelength transmission filter and a first long wavelength transmission filter of the plurality of long wavelength transmission filters. The method of claim 1, wherein the method is a difference. Further comprising: identifying a target gas and a contaminated gas; and recording each pixel that exceeds a species identification threshold as a false alarm pixel,
The second active band includes a second difference between a second long wavelength transmission filter of the plurality of long wavelength transmission filters and the reference long wavelength transmission filter,
The step of discriminating between the target gas and the pollutant gas includes a step of comparing a discrimination feature parameter with the species discrimination threshold, wherein the discrimination feature parameter is a ratio ΔL _{1st_active_band} / ΔL _{2nd_active_band} , and ΔL _{1st_active_band} is the first activity. ΔL for the band, ΔL _{2nd_active_band} is ΔL for the second active band, and ΔL is the background radiance L _off estimated in one of the plurality of active bands, and the gas The difference from the radiance L _on detected in the one active band in the presence,
The method of claim 4, wherein the list of alarm conditions further includes each pixel that is not a false alarm pixel having a gas concentration / length greater than a gas threshold. The machine learning is performed on at least one of simulation data and experimental measurement data;
The simulation data is based on a simulation performed based on at least one element selected from the list consisting of known atmospheric temperature, background temperature, background reflectivity, and Planck's law,
The experimental measurement data measures at least one element selected from the list consisting of reflectance of background material, emissivity of background material, background material temperature, atmospheric temperature, temperature difference between the background material and the atmosphere, and sunlight conditions. The method of claim 3, based on an experimental measurement comprising:   The method of claim 1, wherein the target gas is a hydrocarbon gas.   The method of claim 7, wherein the hydrocarbon gas is methane.   The method of claim 5, wherein the polluting gas is water vapor. The method of claim 1, wherein the spectral range is 3 × 10 ⁻⁶ m to 5 × 10 ⁻⁶ m. The method of claim 1, wherein the spectral range is 7 × 10 ⁻⁶ m to 14 × 10 ⁻⁶ m. A multi-spectral optical gas imaging camera, and a processor connected to the multi-spectral optical gas imaging camera,
The processor is
Reading an image containing multiple pixels,
Estimating a background radiance of at least one of the plurality of pixels;
Calculating the gas concentration / length of the at least one of the plurality of pixels based on the image and the background radiance; and if the gas concentration / length of at least one pixel exceeds a gas threshold level, an alarm is generated. To do
The multispectral configuration of the multispectral optical gas imaging camera includes a reference band that is outside an absorption wavelength range of a target gas and an active band that includes at least a portion of the absorption wavelength range of the target gas;
Estimating the background radiance is:
Determining a model relating the detected radiance of the active band to the detected radiance of the reference band; and using the model as a calibration model, the background radiance of the active band in the presence or absence of the target gas Gas concentration and length quantification system, including estimating   The system of claim 12, wherein the image is an infrared image and is acquired by the multispectral optical gas imaging camera capable of sensing infrared radiation.   The system of claim 12, wherein the model is an analytical model determined using machine learning including regression and neural networks. The multispectral configuration includes a plurality of active bands;
The plurality of active bands and the reference band are determined using a plurality of long wavelength transmission filters, one more than the number of active bands,
Each of the plurality of active bands includes the reference band;
The plurality of active bands include a first active band, and the first active band includes a first long wavelength transmission filter and a first long wavelength transmission filter of the plurality of long wavelength transmission filters. The system of claim 12, wherein the system is a difference. The processor further includes:
Identify the target gas and pollutant gas, and record each pixel that exceeds the species identification threshold as a false alarm pixel,
The second active band includes a second difference between a second long wavelength transmission filter of the plurality of long wavelength transmission filters and the reference long wavelength transmission filter,
Discriminating between the target gas and the contaminated gas includes comparing an identification feature parameter to the species identification threshold, wherein the identification feature parameter is a ratio ΔL _{1st_active_band} / ΔL _{2nd_active_band} , where ΔL _{1st_active_band} is the first activity. ΔL for the band, ΔL _{2nd_active_band} is ΔL for the second active band, and ΔL is the background radiance L _off estimated in one of the plurality of active bands, and the gas The difference from the radiance L _on detected in the one active band in the presence,
The system of claim 15, wherein the list of alarm conditions further includes each pixel that is not a false alarm pixel having a gas concentration / length greater than a gas threshold. The machine learning is performed on at least one of simulation data and experimental measurement data;
The simulation data is based on a simulation performed based on at least one element selected from the list consisting of known atmospheric temperature, background temperature, background reflectivity, and Planck's law,
The experimental measurement data measures at least one element selected from the list consisting of reflectance of background material, emissivity of background material, background material temperature, atmospheric temperature, temperature difference between the background material and the atmosphere, and sunlight conditions. 15. The system of claim 14, based on an experimental measurement that includes: A non-transitory computer readable medium having stored therein computer readable program code embodied therein, the computer readable program code comprising:
Read an image containing multiple pixels,
Estimating a background radiance of at least one of the plurality of pixels;
Calculate the gas concentration / length of the at least one of the plurality of pixels based on the image and the background radiance, and issue an alarm if the gas concentration / length of at least one pixel exceeds a gas threshold level ,
The multispectral configuration of the multispectral optical gas imaging camera includes a reference band that is outside an absorption wavelength range of a target gas and an active band that includes at least a portion of the absorption wavelength range of the target gas;
Estimating the background radiance is:
Determining a model relating the detected radiance of the active band to the detected radiance of the reference band; and using the model as a calibration model, the background radiance of the active band in the presence or absence of the target gas Including estimating the medium.   The non-transitory computer readable medium of claim 18, wherein the image is an infrared image and is acquired by an optical gas imaging camera capable of sensing infrared radiation.   The non-transitory computer readable medium of claim 18, wherein the model is an analytical model determined using machine learning including regression and neural networks. The multispectral configuration includes a plurality of active bands;
The plurality of active bands and the reference band are determined using a plurality of long wavelength transmission filters, one more than the number of active bands,
Each of the plurality of active bands includes the reference band;
The plurality of active bands include a first active band, and the first active band includes a first long wavelength transmission filter and a first long wavelength transmission filter of the plurality of long wavelength transmission filters. The non-transitory computer readable medium of claim 18, wherein the difference is a difference. The program code further includes:
Each pixel that distinguishes the target gas from the pollutant gas and exceeds the species identification threshold is recorded as a false alarm pixel,
The second active band includes a second difference between a second long wavelength transmission filter of the plurality of long wavelength transmission filters and the reference long wavelength transmission filter,
Discriminating between the target gas and the contaminated gas includes comparing an identification feature parameter to the species identification threshold, wherein the identification feature parameter is a ratio ΔL _{1st_active_band} / ΔL _{2nd_active_band} , where ΔL _{1st_active_band} is the first activity. ΔL for the band, ΔL _{2nd_active_band} is ΔL for the second active band, and ΔL is the background radiance L _off estimated in one of the plurality of active bands, and the gas The difference from the radiance L _on detected in the one active band in the presence,
The non-transitory computer readable medium of claim 21, wherein the list of alarm conditions further includes each pixel that is not a false alarm pixel having a gas concentration / length greater than a gas threshold. The machine learning is performed on at least one of simulation data and experimental measurement data;
The simulation data is based on a simulation performed based on at least one element selected from the list consisting of known atmospheric temperature, background temperature, background reflectivity, and Planck's law,
The experimental measurement data measures at least one element selected from the list consisting of reflectance of background material, emissivity of background material, background material temperature, atmospheric temperature, temperature difference between the background material and the atmosphere, and sunlight conditions. 21. The non-transitory computer readable medium of claim 20, based on an experimental measurement comprising: